LMR14020: Sub-harmonic instability.

Chris Hickam

Part Number: LMR14020

I put together a design using the LMR14020 which I believe to be unstable. I simulated the design with the PSpice AC and transient models provided on TI's website. According to the averaged model the phase margin is 59d and the gain margin is 45dB. The design worked beautifully in the transient simulation as well for line and load steps. Upon building a prototype and testing it, I found the converter will not regulate at a duty cycle above 50%. Any suggestions? At present I am looking for a different chip with which I can implement my own compensation.

over 7 years ago

0 Vental Mao over 7 years ago

TI__Expert 5745 points

Can you show me you schematic?

and also your waveform about SW,Vin, Vout, IL(all in same screen)

0 Chris Hickam over 7 years ago in reply to Vental Mao

Intellectual 260 points

Hello Vental,

I can't send a schematic, but it is pretty conventional. The input capacitors are, a 47uF tantalum, a 2.2uF ceramic and a .1uF ceramic. Cboot is an .1uF ceramic, the inductor is 22uH, the output capacitors are two low ESR 10uF ceramics. The load has a 330uF electrolytic capacitor built in. I'm switching at 1.2MHz. Whether or not I am connected to the capacitive load I am having this problem. I have captured some waveforms which I will attach. I've tried both increasing and reducing the inductance neither helped.

At Vin = 31V, the duty cycle is well under 50% so no problem. The converter works great at low D.

At 28V, D starts getting closer to 50% and the converter will no longer regulate.

At 20V.

At 16V

Thanks,

Chris

0 Chris Hickam over 7 years ago in reply to Chris Hickam

Intellectual 260 points

0 Chris Hickam over 7 years ago in reply to Chris Hickam

Intellectual 260 points

The order is reversed on the images, the bottom one is 31V in and the top is 16Vin. Since this seems to be what TI's support is now maybe this site could use a little work so customers don't have to reply 3 times.

0 Chris Hickam over 7 years ago in reply to Chris Hickam

Intellectual 260 points

Thanks for your help Vental. I assume in the absence of a response that there is nothing that can be done and I need to find a different controller.

0 Vental Mao over 7 years ago in reply to Chris Hickam

TI__Expert 5745 points

Sorry for the later, we are in Chinese new year holiday.
Do you still use LMR14020?

0 Chris Hickam over 7 years ago in reply to Vental Mao

Intellectual 260 points

I've went to a device which I can design the compensator for.

Power management

Power management forum

LMR14020: Sub-harmonic instability.